I suggest you clarify what exactly you want to achieve and share concrete information log files, otherwise we all have to guess.
BTW, why do you expect "similar speed" with two runs using half the CU resources and sharing a GPU compared to using a single run that does not share the GPU (nor CPU with another)? -- Szilárd On Thu, Nov 19, 2015 at 3:45 AM, RJ <ra...@kaist.ac.kr> wrote: > Hi there, > > > I have 24 threads in my PC with one GTX 980Ti GPU. I would like to run two > simulation job by assigning 12 threads for each job. I have tried using > "-ntomp 12 -ntmpi 1" for mentioning the threads and gpu. However, i couldnt > get the similar speed as if run them alone with 12 threads. I have even > used "-pin on" but haven't get any significant changes in the speed? > > > Its much appreciated if you could provide the hints on it. Thanks a lot. > -- > Gromacs Users mailing list > > * Please search the archive at > http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before > posting! > > * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists > > * For (un)subscribe requests visit > https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or > send a mail to gmx-users-requ...@gromacs.org. > -- Gromacs Users mailing list * Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting! * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org.
